对康威-罗伯茨文章的一些观察、一个建议及若干评论

Some Observations, a Suggestion, and Some Comments on the Conway-Roberts Article

Journal of Business & Economic Statistics · 1984
被引 1
人大 AABS 4

中文导读

指出在两种常见情形下,同时存在对女性和男性的不公平(按康威-罗伯茨的术语)在数学上必然意味着男性最初处于另一种不公平地位,并建议使用符号检验作为替代分析方法,适用于比较两组数值度量且存在多个分层变量的情况。

Abstract

This comment has two purposes. First, it will be shown that in two common situations, whenever one has unfairness 1 (using Conway and Roberts's terminology) against females and unfairness against males simultaneously, it is a mathematical necessity that males must have been in a position that was unfair in sense 2 (to use Conway and Roberts's phrase) at the outset. This mathematical result leads to an interesting question of semantics. Second, we will suggest an alternate method of analysis that can be used in situations similar to those described in the Conway and Roberts article involving qualifications and also in many other situations. The method is particularly useful whenever two groups are being compared on a numerical measure and there are several potential stratifying variables. The technique we shall propose involves use of the sign test, a nonparametric method easily explained to and understood by lawyers, judges, corporate executives, and other interested parties.

性别歧视不公平性符号检验非参数方法